import os
import numpy as np
import pandas as pd 
import matplotlib.pyplot as plt
import seaborn as sns
from scipy.stats import skew
from sklearn.preprocessing import PowerTransformer, RobustScaler

df = pd.read_csv("american_bankruptcy.csv")
df.head()

df.info()
df.describe()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 78682 entries, 0 to 78681
Data columns (total 21 columns):
 #   Column        Non-Null Count  Dtype  
---  ------        --------------  -----  
 0   company_name  78682 non-null  object 
 1   status_label  78682 non-null  object 
 2   year          78682 non-null  int64  
 3   X1            78682 non-null  float64
 4   X2            78682 non-null  float64
 5   X3            78682 non-null  float64
 6   X4            78682 non-null  float64
 7   X5            78682 non-null  float64
 8   X6            78682 non-null  float64
 9   X7            78682 non-null  float64
 10  X8            78682 non-null  float64
 11  X9            78682 non-null  float64
 12  X10           78682 non-null  float64
 13  X11           78682 non-null  float64
 14  X12           78682 non-null  float64
 15  X13           78682 non-null  float64
 16  X14           78682 non-null  float64
 17  X15           78682 non-null  float64
 18  X16           78682 non-null  float64
 19  X17           78682 non-null  float64
 20  X18           78682 non-null  float64
dtypes: float64(18), int64(1), object(2)
memory usage: 12.6+ MB

df = df.rename( columns={
'X1':'Current assets',
'X2':'Cost of goods sold',
'X3': 'Depreciation and amortization',
'X4':'EBITDA',
'X5': 'Inventory',
'X6' : "Net Income",
'X7':'Total Receivables',
'X8' : 'Market Value',
'X9' : 'Net sales',
'X10': 'Total assets',
'X11': 'Total Long-term debt',
'X12' : 'EBIT',
'X13':'Gross Profit',
'X14': 'Total Current Liabilities',
'X15' : 'Retained Earnings',
'X16': 'Total Revenue',
'X17' : 'Total Liabilities',
'X18': 'Total Operating Expenses',

})
df.columns

Index(['company_name', 'status_label', 'year', 'Current assets',
       'Cost of goods sold', 'Depreciation and amortization', 'EBITDA',
       'Inventory', 'Net Income', 'Total Receivables', 'Market Value',
       'Net sales', 'Total assets', 'Total Long-term debt', 'EBIT',
       'Gross Profit', 'Total Current Liabilities', 'Retained Earnings',
       'Total Revenue', 'Total Liabilities', 'Total Operating Expenses'],
      dtype='object')

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 78682 entries, 0 to 78681
Data columns (total 21 columns):
 #   Column                         Non-Null Count  Dtype  
---  ------                         --------------  -----  
 0   company_name                   78682 non-null  object 
 1   status_label                   78682 non-null  object 
 2   year                           78682 non-null  int64  
 3   Current assets                 78682 non-null  float64
 4   Cost of goods sold             78682 non-null  float64
 5   Depreciation and amortization  78682 non-null  float64
 6   EBITDA                         78682 non-null  float64
 7   Inventory                      78682 non-null  float64
 8   Net Income                     78682 non-null  float64
 9   Total Receivables              78682 non-null  float64
 10  Market Value                   78682 non-null  float64
 11  Net sales                      78682 non-null  float64
 12  Total assets                   78682 non-null  float64
 13  Total Long-term debt           78682 non-null  float64
 14  EBIT                           78682 non-null  float64
 15  Gross Profit                   78682 non-null  float64
 16  Total Current Liabilities      78682 non-null  float64
 17  Retained Earnings              78682 non-null  float64
 18  Total Revenue                  78682 non-null  float64
 19  Total Liabilities              78682 non-null  float64
 20  Total Operating Expenses       78682 non-null  float64
dtypes: float64(18), int64(1), object(2)
memory usage: 12.6+ MB

Null_value = [x for x in df.columns if df[x].isnull().sum()>0]
for x in Null_value:
    print(x, df[x].isnull().sum())
else:
    print('No Missisng Values')

No Missisng Values

for i in df.columns:
    print(f"{i}:, {df[i].nunique()}")

company_name:, 8971
status_label:, 2
year:, 20
Current assets:, 65895
Cost of goods sold:, 65690
Depreciation and amortization:, 36010
EBITDA:, 59060
Inventory:, 38898
Net Income:, 55550
Total Receivables:, 49577
Market Value:, 77580
Net sales:, 68596
Total assets:, 71521
Total Long-term debt:, 39741
EBIT:, 56949
Gross Profit:, 64952
Total Current Liabilities:, 58685
Retained Earnings:, 72062
Total Revenue:, 68596
Total Liabilities:, 64640
Total Operating Expenses:, 70840

df['status_label'].astype('str').str.lower()
df['status_label'] = df['status_label'] = df['status_label'].replace({
    'alive': 1,
    'failed': 0
}).astype('category')
df['status_label'].nunique()

C:\Users\AFC\AppData\Local\Temp\ipykernel_2788\240093521.py:2: FutureWarning: Downcasting behavior in `replace` is deprecated and will be removed in a future version. To retain the old behavior, explicitly call `result.infer_objects(copy=False)`. To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)`
  df['status_label'] = df['status_label'] = df['status_label'].replace({

2

df[['company_name','year']].duplicated().sum()

np.int64(0)

categorical_features = df.select_dtypes(include=['category']).columns.tolist()
categorical_features
df.head()

# list of numerical variables
numerical_features = df.select_dtypes(include=['int64', 'float64']).columns.tolist()
print('Number of numerical variables: ', len(numerical_features))

# visualise the numerical variables
df[numerical_features].nunique()

Number of numerical variables:  19

year                                20
Current assets                   65895
Cost of goods sold               65690
Depreciation and amortization    36010
EBITDA                           59060
Inventory                        38898
Net Income                       55550
Total Receivables                49577
Market Value                     77580
Net sales                        68596
Total assets                     71521
Total Long-term debt             39741
EBIT                             56949
Gross Profit                     64952
Total Current Liabilities        58685
Retained Earnings                72062
Total Revenue                    68596
Total Liabilities                64640
Total Operating Expenses         70840
dtype: int64

discrete_features = [col for col in numerical_features if df[col].nunique() <= 30]
continuous_features = [col for col in numerical_features if df[col].nunique() > 30]
discrete_features = df[discrete_features].astype('int')
# Step 3: Print results
print("Number of numerical columns:", len(numerical_features))
print("Discrete Features:", discrete_features, "Data type", discrete_features.dtypes)
print("Continuous Features:", continuous_features)

Number of numerical columns: 19
Discrete Features:        year
0      1999
1      2000
2      2001
3      2002
4      2003
...     ...
78677  2014
78678  2015
78679  2016
78680  2017
78681  2018

[78682 rows x 1 columns] Data type year    int64
dtype: object
Continuous Features: ['Current assets', 'Cost of goods sold', 'Depreciation and amortization', 'EBITDA', 'Inventory', 'Net Income', 'Total Receivables', 'Market Value', 'Net sales', 'Total assets', 'Total Long-term debt', 'EBIT', 'Gross Profit', 'Total Current Liabilities', 'Retained Earnings', 'Total Revenue', 'Total Liabilities', 'Total Operating Expenses']

def plot_categotical_distribution(df,categorical_features):
    plt.figure(figsize = (6,4))
    for i, col in enumerate(categorical_features):
        value_counts = df[col].value_counts()
        sns.barplot(x = value_counts.index, y=value_counts.values)
        plt.title(f"Distribution of {col}")
        plt.xlabel(col)
        plt.ylabel("Counts")
    plt.show
plot_categotical_distribution(df,categorical_features)

def plot_discrete_features(df, discrete_features):
    plt.figure(figsize = (6,4))
    for i, col in enumerate(discrete_features):
        sns.histplot(data = discrete_features, x = col)
        plt.title(f"Distribution of {col}")
        plt.xlabel(col)
        plt.ylabel('Count')
    plt.tight_layout()
    plt.show()
plot_discrete_features(df, discrete_features)

df[continuous_features].describe()

for col in continuous_features:
    plt.figure(figsize=(12, 4))
    plt.subplot(1, 2, 1)
    sns.histplot(df[col], kde=True)
    plt.title(f"Histogram of {col}")
    
    sk = df[col].skew()
    plt.title(f'{col}\nSkewness: {sk:.2f}')

    plt.subplot(1, 2, 2)
    sns.boxplot(x=df[col])
    plt.title(f"Boxplot of {col}")
    plt.tight_layout()
    plt.show()

def eda_transform_plot(df, num, skew_threshold=0.5, return_transformed=False):
    pt = PowerTransformer(method='yeo-johnson')
    scaler = RobustScaler()
    
    transformed_data = {}

    for col in num:
        data = df[col].dropna()
        skewness = skew(data)
        print(f"{col} - Skewness: {skewness:.2f}")

        if abs(skewness) < skew_threshold:
            print(f"→ Skew is low. Skipping transformation for: {col}")
            transformed_data[col] = data
            continue

        # Transform
        reshaped = data.values.reshape(-1, 1)
        data_pt = pt.fit_transform(reshaped)
        data_scaled = scaler.fit_transform(data_pt).flatten()
        transformed_data[col] = pd.Series(data_scaled, index=data.index)

        # Plot
        plt.figure(figsize=(12, 4))

        plt.subplot(1, 2, 1)
        sns.histplot(data, kde=True, bins=30, color='steelblue')
        plt.title(f"Original: {col}\nSkew: {skewness:.2f}")

        plt.subplot(1, 2, 2)
        sns.histplot(data_scaled, kde=True, bins=30, color='seagreen')
        new_skew = skew(data_scaled)
        plt.title(f"Transformed: {col}\nSkew: {new_skew:.2f}")

        plt.tight_layout()
        plt.show()
    
    if return_transformed:
        return pd.DataFrame(transformed_data)

# To just visualize
eda_transform_plot(df, continuous_features)

# To get transformed values
continuous_features = eda_transform_plot(df, continuous_features, return_transformed=True)

Current assets - Skewness: 14.84

Cost of goods sold - Skewness: 20.05

Depreciation and amortization - Skewness: 17.86

EBITDA - Skewness: 16.40

Inventory - Skewness: 22.57

Net Income - Skewness: 11.87

Total Receivables - Skewness: 15.84

Market Value - Skewness: 18.19

Net sales - Skewness: 18.99

Total assets - Skewness: 13.56

continuous_features = df.select_dtypes(include=['float64', 'int64']).columns.tolist()
df[continuous_features]

print("Continious Features : ",continuous_features
      ,'\nDiscrete Feature: ',discrete_features.columns,'\nCategorical Features: ',categorical_features)

Continious Features :  ['year', 'Current assets', 'Cost of goods sold', 'Depreciation and amortization', 'EBITDA', 'Inventory', 'Net Income', 'Total Receivables', 'Market Value', 'Net sales', 'Total assets', 'Total Long-term debt', 'EBIT', 'Gross Profit', 'Total Current Liabilities', 'Retained Earnings', 'Total Revenue', 'Total Liabilities', 'Total Operating Expenses'] 
Discrete Feature:  Index(['year'], dtype='object') 
Categorical Features:  ['status_label']

corr_matrix = df[continuous_features].corr(method='pearson')
plt.figure(figsize=(14,10))
sns.heatmap(corr_matrix, annot = True, cmap='coolwarm')
plt.title("Pearson Correlation Matrix")
plt.show()

# Set your correlation threshold
correlation_threshold = 0.85

# Step 1: Compute Pearson correlation matrix
corr_matrix = df[continuous_features].corr().abs()

# Step 2: Select upper triangle of correlation matrix (to avoid duplicates)
upper_tri = corr_matrix.where(np.triu(np.ones(corr_matrix.shape), k=1).astype(bool))

# Step 3: Find highly correlated feature pairs
to_drop = []
suggestions = []

for col in upper_tri.columns:
    high_corr_features = upper_tri.index[upper_tri[col] > correlation_threshold].tolist()
    
    for correlated_col in high_corr_features:
        # Step 4: Compare average correlations of both features
        avg_corr_col = corr_matrix[col].mean()
        avg_corr_correlated = corr_matrix[correlated_col].mean()

        # Drop the one with higher average correlation (more redundant)
        if avg_corr_col > avg_corr_correlated:
            drop_feature = col
            keep_feature = correlated_col
        else:
            drop_feature = correlated_col
            keep_feature = col

        if drop_feature not in to_drop:
            to_drop.append(drop_feature)
            suggestions.append({
                "Drop": drop_feature,
                "Keep": keep_feature,
                "Reason": f"{drop_feature} is more correlated with other features (avg corr: {avg_corr_col:.2f} vs {avg_corr_correlated:.2f})"
            })

# Step 5: Output recommendations
suggestions_df = pd.DataFrame(suggestions)
print("🔍 Suggested Features to Remove Based on Correlation:\n")
print(suggestions_df)
print(to_drop)

🔍 Suggested Features to Remove Based on Correlation:

                        Drop                           Keep  \
0                     EBITDA  Depreciation and amortization   
1             Current assets              Total Receivables   
2                  Net sales             Cost of goods sold   
3               Total assets  Depreciation and amortization   
4                       EBIT                   Market Value   
5               Gross Profit                         EBITDA   
6  Total Current Liabilities                 Current assets   
7              Total Revenue             Cost of goods sold   
8          Total Liabilities  Depreciation and amortization   
9   Total Operating Expenses             Cost of goods sold   

                                              Reason  
0  EBITDA is more correlated with other features ...  
1  Current assets is more correlated with other f...  
2  Net sales is more correlated with other featur...  
3  Total assets is more correlated with other fea...  
4  EBIT is more correlated with other features (a...  
5  Gross Profit is more correlated with other fea...  
6  Total Current Liabilities is more correlated w...  
7  Total Revenue is more correlated with other fe...  
8  Total Liabilities is more correlated with othe...  
9  Total Operating Expenses is more correlated wi...  
['EBITDA', 'Current assets', 'Net sales', 'Total assets', 'EBIT', 'Gross Profit', 'Total Current Liabilities', 'Total Revenue', 'Total Liabilities', 'Total Operating Expenses']

# Step 4: Drop the columns from the DataFrame
reduced_continuous_features = df[continuous_features].drop(columns=to_drop)
reduced_continuous_features.columns.to_list()

['year',
 'Cost of goods sold',
 'Depreciation and amortization',
 'Inventory',
 'Net Income',
 'Total Receivables',
 'Market Value',
 'Total Long-term debt',
 'Retained Earnings']

plt.figure(figsize = (10,8))
sns.heatmap(reduced_continuous_features.corr(method = 'pearson'), annot = True, cmap='coolwarm')
plt.title('Reduced Continuous Features Heatmap')
plt.show()

sns.pairplot(reduced_continuous_features.sample(300), corner=True)
plt.show()

from scipy.stats import ttest_ind

for col in reduced_continuous_features:
    # Drop NA values
    temp = df[[col, 'status_label']].dropna()
    group1 = temp[temp['status_label'] == 0][col]
    group2 = temp[temp['status_label'] == 1][col]

    # Perform t-test
    t_stat, p_val = ttest_ind(group1, group2, equal_var=False)
    print(f"{col} → t-stat: {t_stat:.2f}, p-value: {p_val:.4f}")

year → t-stat: -27.53, p-value: 0.0000
Cost of goods sold → t-stat: -15.44, p-value: 0.0000
Depreciation and amortization → t-stat: -8.48, p-value: 0.0000
Inventory → t-stat: -13.45, p-value: 0.0000
Net Income → t-stat: -20.02, p-value: 0.0000
Total Receivables → t-stat: -21.35, p-value: 0.0000
Market Value → t-stat: -32.42, p-value: 0.0000
Total Long-term debt → t-stat: -3.88, p-value: 0.0001
Retained Earnings → t-stat: -21.19, p-value: 0.0000

for col in reduced_continuous_features.columns:
    plt.figure(figsize=(4, 4))
    sns.barplot(x='status_label', y=col, data=df)
    plt.title(f'{col} vs. status_label')
    plt.tight_layout()
    plt.show()

from scipy.stats import chi2_contingency  # Required import

pd_crosstab = pd.crosstab(df['year'], df['status_label'])
chi2, p, dof, expected = chi2_contingency(pd_crosstab)
print("Chi-Square Test Results")
print("------------------------")
print("Chi2 Statistic :", round(chi2, 4))
print("p-value        :", round(p, 4))
print("Degrees of Freedom :", dof)
print("\nExpected Frequencies:")
expected_df = pd.DataFrame(expected, index=pd_crosstab.index, columns=pd_crosstab.columns)
print(expected_df)

Chi-Square Test Results
------------------------
Chi2 Statistic : 731.2001
p-value        : 0.0
Degrees of Freedom : 19

Expected Frequencies:
status_label           0            1
year                                 
1999          352.148649  4955.851351
2000          346.708523  4879.291477
2001          324.881676  4572.118324
2002          308.561297  4342.438703
2003          293.037035  4123.962965
2004          288.459368  4059.540632
2005          278.972319  3926.027681
2006          273.863908  3854.136092
2007          265.969091  3743.030909
2008          255.884955  3601.115045
2009          248.321853  3494.678147
2010          240.493378  3384.506622
2011          233.062962  3279.937038
2012          231.205358  3253.794642
2013          231.139015  3252.860985
2014          232.465875  3271.534125
2015          222.514425  3131.485575
2016          211.700516  2979.299484
2017          199.957805  2814.042195
2018          180.651992  2542.348008

plt.figure(figsize=(12,6))
pd_crosstab.plot(kind='bar', stacked=True, colormap='Set2')
plt.title("Observed Bankruptcy Status by Year")
plt.ylabel("Company Count")
plt.xlabel("Year")
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

<Figure size 1200x600 with 0 Axes>

from statsmodels.stats.outliers_influence import variance_inflation_factor

vif_data = pd.DataFrame()
vif_data["Feature"] = reduced_continuous_features.columns
vif_data["VIF"] = [variance_inflation_factor(reduced_continuous_features.values, i) for i in range(reduced_continuous_features.shape[1])]

# Display VIF sorted by highest first
print(vif_data.sort_values(by="VIF", ascending=False))

                         Feature       VIF
2  Depreciation and amortization  4.376951
1             Cost of goods sold  4.090087
7           Total Long-term debt  3.502274
6                   Market Value  3.191790
5              Total Receivables  3.038389
3                      Inventory  2.869585
8              Retained Earnings  2.551365
4                     Net Income  2.379139
0                           year  1.072124

vif_data_sorted = vif_data.sort_values(by="VIF", ascending=False)

# Horizontal bar plot
plt.figure(figsize=(10, 6))
sns.barplot(x="VIF", y="Feature", data=vif_data_sorted, palette="Blues_r")

plt.title("VIF Scores for Features")
plt.xlabel("Variance Inflation Factor (VIF)")
plt.ylabel("Features")
plt.xlim(0, max(vif_data_sorted["VIF"]) + 1)
plt.grid(axis='x', linestyle='--', alpha=0.5)
plt.show()

C:\Users\AFC\AppData\Local\Temp\ipykernel_2788\1967184281.py:5: FutureWarning: 

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `y` variable to `hue` and set `legend=False` for the same effect.

  sns.barplot(x="VIF", y="Feature", data=vif_data_sorted, palette="Blues_r")

reduced_feature_list = df[['year',
 'Cost of goods sold',
 'Depreciation and amortization',
 'Inventory',
 'Net Income',
 'Total Receivables',
 'Market Value',
 'Total Long-term debt',
 'Retained Earnings']
].columns.tolist()

reduced_df = df[reduced_feature_list + ['status_label']].copy()

# Convert to int if needed
reduced_df['status_label'] = reduced_df['status_label'].astype(int)

# Pairplot
sns.pairplot(reduced_df, hue='status_label', diag_kind='kde')
plt.show()

from scipy.stats import kruskal
df_clean = df.dropna(subset=['status_label'])
df_clean['status_label'] = df_clean['status_label'].astype(int)

for feature in reduced_continuous_features:
    group0 = df_clean[df_clean['status_label'] == 0][feature]
    group1 = df_clean[df_clean['status_label'] == 1][feature]
    
    if len(group0) > 0 and len(group1) > 0:
        stat, p = kruskal(group0, group1)
        print(f"Kruskal-Wallis test for {feature}: H-statistic = {stat:.4f}, p-value = {p:.4e}")
    else:
        print(f"Not enough data in one of the groups for feature: {feature}")

Kruskal-Wallis test for year: H-statistic = 502.3647, p-value = 2.9071e-111
Kruskal-Wallis test for Cost of goods sold: H-statistic = 8.0662, p-value = 4.5099e-03
Kruskal-Wallis test for Depreciation and amortization: H-statistic = 0.3790, p-value = 5.3812e-01
Kruskal-Wallis test for Inventory: H-statistic = 13.0289, p-value = 3.0672e-04
Kruskal-Wallis test for Net Income: H-statistic = 980.3695, p-value = 3.3211e-215
Kruskal-Wallis test for Total Receivables: H-statistic = 133.0323, p-value = 8.8953e-31
Kruskal-Wallis test for Market Value: H-statistic = 384.3159, p-value = 1.4300e-85
Kruskal-Wallis test for Total Long-term debt: H-statistic = 89.8555, p-value = 2.5620e-21
Kruskal-Wallis test for Retained Earnings: H-statistic = 618.1158, p-value = 1.9211e-136

p_values = {}

for feature in reduced_continuous_features:
    group_0 = df_clean[df_clean['status_label'] == 0][feature]
    group_1 = df_clean[df_clean['status_label'] == 1][feature]
    
    # Ensure we have data in both groups
    if len(group_0) > 0 and len(group_1) > 0:
        stat, p = kruskal(group_0, group_1)
        p_values[feature] = p
    else:
        p_values[feature] = None
pval_df = pd.DataFrame.from_dict(p_values, orient='index', columns=['p_value'])

# Apply -log10 to highlight small p-values in heatmap
pval_df['-log10(p)'] = -np.log10(pval_df['p_value'])

# Plot heatmap
plt.figure(figsize=(8, 6))
sns.heatmap(
    pval_df[['-log10(p)']],
    annot=True,
    cmap='coolwarm',
    cbar_kws={'label': '-log10(p-value)'},
    vmin=0, vmax=5,  # sets fixed scale for meaningful comparison
    linewidths=0.5,
    linecolor='green'
)

plt.title('Kruskal-Wallis Test Significance Heatmap')
plt.tight_layout()
plt.show()

plt.figure(figsize=(14, 5))
for i, col in enumerate(['Net Income', 'Market Value'], start=1):
    plt.subplot(1, 2, i)
    sns.barplot(x='status_label', y=col, data=df_clean)
    plt.title(f'{col} vs Status Label')
plt.tight_layout()
plt.show()

status_year = df_clean.groupby(['year', 'status_label']).size().unstack()
status_year.plot(kind='bar', stacked=True, colormap='viridis')
plt.title("Distribution of Status Labels Over Years")
plt.xlabel("Year")
plt.ylabel("Company Count")
plt.tight_layout()
plt.show()

pvals = {}
for feature in reduced_df.columns:
    group0 = df_clean[df_clean['status_label'] == 0][feature]
    group1 = df_clean[df_clean['status_label'] == 1][feature]
    if len(group0) > 0 and len(group1) > 0:
        stat, p = kruskal(group0, group1)
        pvals[feature] = p
    else:
        pvals[feature] = np.nan

pval_df = pd.DataFrame.from_dict(pvals, orient='index', columns=['p_value'])
pval_df['-log10(p)'] = -np.log10(pval_df['p_value'])

# Heatmap of Kruskal significance
plt.figure(figsize=(8, 6))
sns.heatmap(pval_df[['-log10(p)']], annot=True, cmap='coolwarm', cbar_kws={'label': '-log10(p-value)'}, vmin=0, vmax=5)
plt.title('Kruskal-Wallis Test Heatmap (Feature Significance)')
plt.tight_layout()
plt.show()

C:\Python\Lib\site-packages\pandas\core\arraylike.py:399: RuntimeWarning: divide by zero encountered in log10
  result = getattr(ufunc, method)(*inputs, **kwargs)

plt.figure(figsize=(10, 8))
sns.heatmap(reduced_continuous_features.corr(), annot=True, cmap='vlag', center=0)
plt.title("Correlation Matrix of Financial Features")
plt.tight_layout()
plt.show()

numeric_cols = df.select_dtypes(include=[np.number]).columns

grouped = df.groupby('status_label')[numeric_cols].mean().T


grouped.columns = ['Status 0 Avg', 'Status 1 Avg']

grouped['Difference'] = grouped['Status 1 Avg'] - grouped['Status 0 Avg']

print("\n--- Financial Feature Differences Between Groups ---\n")
print(grouped.sort_values('Difference', ascending=False))

grouped['Difference'].sort_values().plot(kind='barh', figsize=(8, 6), color='teal')
plt.axvline(0, color='black', linestyle='--')
plt.title("Mean Feature Differences (Status 1 - Status 0)")
plt.tight_layout()
plt.show()

C:\Users\AFC\AppData\Local\Temp\ipykernel_2788\1258145425.py:3: FutureWarning: The default of observed=False is deprecated and will be changed to True in a future version of pandas. Pass observed=False to retain current behavior or observed=True to adopt the future default and silence this warning.
  grouped = df.groupby('status_label')[numeric_cols].mean().T

--- Financial Feature Differences Between Groups ---

                               Status 0 Avg  Status 1 Avg   Difference
Market Value                     857.813011   3596.015157  2738.202146
Total assets                    1664.630981   2952.555391  1287.924410
Net sales                       1183.957172   2447.871581  1263.914409
Total Revenue                   1183.957172   2447.871581  1263.914409
Total Operating Expenses        1045.388596   2054.187022  1008.798426
Cost of goods sold               856.340997   1646.982563   790.641566
Retained Earnings               -165.107267    582.034710   747.141977
Total Liabilities               1266.816748   1809.571974   542.755226
Current assets                   399.339353    914.542615   515.203262
Gross Profit                     327.616175    800.889131   473.272956
Total Current Liabilities        347.233954    628.748794   281.514840
EBITDA                           138.568576    393.684586   255.116009
EBIT                              52.687657    269.938080   217.250422
Net Income                       -48.112333    141.994726   190.107058
Total Receivables                132.565307    297.794547   165.229240
Total Long-term debt             609.430007    730.516984   121.086977
Inventory                        103.735398    208.560102   104.824704
Depreciation and amortization     85.880919    123.746364    37.865445
year                            2005.710920   2007.633892     1.922973

	company_name	status_label	year	X1	X2	X3	X4	X5	X6	X7	...	X9	X10	X11	X12	X13	X14	X15	X16	X17	X18
0	C_1	alive	1999	511.267	833.107	18.373	89.031	336.018	35.163	128.348	...	1024.333	740.998	180.447	70.658	191.226	163.816	201.026	1024.333	401.483	935.302
1	C_1	alive	2000	485.856	713.811	18.577	64.367	320.590	18.531	115.187	...	874.255	701.854	179.987	45.790	160.444	125.392	204.065	874.255	361.642	809.888
2	C_1	alive	2001	436.656	526.477	22.496	27.207	286.588	-58.939	77.528	...	638.721	710.199	217.699	4.711	112.244	150.464	139.603	638.721	399.964	611.514
3	C_1	alive	2002	396.412	496.747	27.172	30.745	259.954	-12.410	66.322	...	606.337	686.621	164.658	3.573	109.590	203.575	124.106	606.337	391.633	575.592
4	C_1	alive	2003	432.204	523.302	26.680	47.491	247.245	3.504	104.661	...	651.958	709.292	248.666	20.811	128.656	131.261	131.884	651.958	407.608	604.467

	year	X1	X2	X3	X4	X5	X6	X7	X8	X9	X10	X11	X12	X13	X14	X15	X16	X17	X18
count	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	7.868200e+04	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000
mean	2007.506317	880.362485	1594.529029	121.234256	376.759424	201.605717	129.382453	286.832743	3.414355e+03	2364.019706	2867.110620	722.483710	255.525035	769.490783	610.072255	532.467069	2364.019706	1773.563963	1987.260307
std	5.742768	3928.564794	8930.484664	652.376804	2012.023142	1060.766096	1265.532022	1335.978571	1.841410e+04	11950.068842	12917.944421	3242.170946	1494.643534	3774.703114	2938.387443	6369.159440	11950.068842	8053.684902	10419.629038
min	1999.000000	-7.760000	-366.645000	0.000000	-21913.000000	0.000000	-98696.000000	-0.006000	1.000000e-04	-1964.999000	0.001000	-0.023000	-25913.000000	-21536.000000	0.001000	-102362.000000	-1964.999000	0.001000	-317.197000
25%	2002.000000	18.924000	17.038250	1.192000	-0.811000	0.000000	-7.415750	3.281250	3.498000e+01	27.548500	37.363500	0.000000	-2.787000	8.521250	8.889250	-68.282750	27.548500	13.486000	32.872500
50%	2007.000000	100.449500	103.661000	7.929500	15.034500	7.023000	1.616000	22.820000	2.275118e+02	186.598500	213.203500	7.593500	6.518000	63.581500	43.333000	-1.131000	186.598500	81.988000	168.912000
75%	2012.000000	431.526750	634.548000	47.971750	139.655250	74.747250	40.144250	131.580500	1.244890e+03	1046.402500	1171.364750	248.760750	87.599000	344.074250	222.817000	146.070000	1046.402500	629.975000	875.522250
max	2018.000000	169662.000000	374623.000000	28430.000000	81730.000000	62567.000000	104821.000000	65812.000000	1.073391e+06	511729.000000	531864.000000	166250.000000	71230.000000	137106.000000	116866.000000	402089.000000	511729.000000	337980.000000	481580.000000

	company_name	status_label	year	Current assets	Cost of goods sold	Depreciation and amortization	EBITDA	Inventory	Net Income	Total Receivables	...	Net sales	Total assets	Total Long-term debt	EBIT	Gross Profit	Total Current Liabilities	Retained Earnings	Total Revenue	Total Liabilities	Total Operating Expenses
0	C_1	1	1999	511.267	833.107	18.373	89.031	336.018	35.163	128.348	...	1024.333	740.998	180.447	70.658	191.226	163.816	201.026	1024.333	401.483	935.302
1	C_1	1	2000	485.856	713.811	18.577	64.367	320.590	18.531	115.187	...	874.255	701.854	179.987	45.790	160.444	125.392	204.065	874.255	361.642	809.888
2	C_1	1	2001	436.656	526.477	22.496	27.207	286.588	-58.939	77.528	...	638.721	710.199	217.699	4.711	112.244	150.464	139.603	638.721	399.964	611.514
3	C_1	1	2002	396.412	496.747	27.172	30.745	259.954	-12.410	66.322	...	606.337	686.621	164.658	3.573	109.590	203.575	124.106	606.337	391.633	575.592
4	C_1	1	2003	432.204	523.302	26.680	47.491	247.245	3.504	104.661	...	651.958	709.292	248.666	20.811	128.656	131.261	131.884	651.958	407.608	604.467

	Current assets	Cost of goods sold	Depreciation and amortization	EBITDA	Inventory	Net Income	Total Receivables	Market Value	Net sales	Total assets	Total Long-term debt	EBIT	Gross Profit	Total Current Liabilities	Retained Earnings	Total Revenue	Total Liabilities	Total Operating Expenses
count	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	7.868200e+04	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000	78682.000000
mean	880.362485	1594.529029	121.234256	376.759424	201.605717	129.382453	286.832743	3.414355e+03	2364.019706	2867.110620	722.483710	255.525035	769.490783	610.072255	532.467069	2364.019706	1773.563963	1987.260307
std	3928.564794	8930.484664	652.376804	2012.023142	1060.766096	1265.532022	1335.978571	1.841410e+04	11950.068842	12917.944421	3242.170946	1494.643534	3774.703114	2938.387443	6369.159440	11950.068842	8053.684902	10419.629038
min	-7.760000	-366.645000	0.000000	-21913.000000	0.000000	-98696.000000	-0.006000	1.000000e-04	-1964.999000	0.001000	-0.023000	-25913.000000	-21536.000000	0.001000	-102362.000000	-1964.999000	0.001000	-317.197000
25%	18.924000	17.038250	1.192000	-0.811000	0.000000	-7.415750	3.281250	3.498000e+01	27.548500	37.363500	0.000000	-2.787000	8.521250	8.889250	-68.282750	27.548500	13.486000	32.872500
50%	100.449500	103.661000	7.929500	15.034500	7.023000	1.616000	22.820000	2.275118e+02	186.598500	213.203500	7.593500	6.518000	63.581500	43.333000	-1.131000	186.598500	81.988000	168.912000
75%	431.526750	634.548000	47.971750	139.655250	74.747250	40.144250	131.580500	1.244890e+03	1046.402500	1171.364750	248.760750	87.599000	344.074250	222.817000	146.070000	1046.402500	629.975000	875.522250
max	169662.000000	374623.000000	28430.000000	81730.000000	62567.000000	104821.000000	65812.000000	1.073391e+06	511729.000000	531864.000000	166250.000000	71230.000000	137106.000000	116866.000000	402089.000000	511729.000000	337980.000000	481580.000000

	year	Current assets	Cost of goods sold	Depreciation and amortization	EBITDA	Inventory	Net Income	Total Receivables	Market Value	Net sales	Total assets	Total Long-term debt	EBIT	Gross Profit	Total Current Liabilities	Retained Earnings	Total Revenue	Total Liabilities	Total Operating Expenses
0	1999	511.267	833.107	18.373	89.031	336.018	35.163	128.348	372.7519	1024.333	740.998	180.447	70.658	191.226	163.816	201.026	1024.333	401.483	935.302
1	2000	485.856	713.811	18.577	64.367	320.590	18.531	115.187	377.1180	874.255	701.854	179.987	45.790	160.444	125.392	204.065	874.255	361.642	809.888
2	2001	436.656	526.477	22.496	27.207	286.588	-58.939	77.528	364.5928	638.721	710.199	217.699	4.711	112.244	150.464	139.603	638.721	399.964	611.514
3	2002	396.412	496.747	27.172	30.745	259.954	-12.410	66.322	143.3295	606.337	686.621	164.658	3.573	109.590	203.575	124.106	606.337	391.633	575.592
4	2003	432.204	523.302	26.680	47.491	247.245	3.504	104.661	308.9071	651.958	709.292	248.666	20.811	128.656	131.261	131.884	651.958	407.608	604.467
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
78677	2014	233.211	43.338	14.094	45.615	3.376	25.261	22.846	756.4827	104.223	1099.101	184.666	31.521	60.885	28.197	28.095	104.223	225.887	58.608
78678	2015	105.559	59.184	42.592	202.133	2.288	129.688	54.611	527.5750	291.153	1865.926	770.103	159.541	231.969	88.128	157.783	291.153	880.327	89.020
78679	2016	63.971	69.074	65.057	79.051	2.581	-1.442	42.467	578.8868	169.858	1746.235	683.985	13.994	100.784	85.765	156.341	169.858	770.233	90.807
78680	2017	135.207	66.527	65.330	69.171	2.013	-20.401	27.217	412.6241	161.884	1736.110	694.035	3.841	95.357	82.010	135.941	161.884	776.697	92.713
78681	2018	82.589	68.817	65.201	67.262	2.112	-50.946	45.839	354.1786	160.513	1625.370	632.122	2.061	91.696	79.365	84.995	160.513	712.687	93.251